As the amount of information on the World Wide Web grows, it becomes increasingly burdensome to and just what we want. While general-purpose search engines such as Ask.com and Bing high coverage, they often provide only low precision compared to others, even for detailed and relative queries. When we know that we want information about a certain type, or on a certain topic, a domain-specific search engine can be a powerful tool. Like www.campsearch.com allows complex queries over summer camps by age-group, size, location, and cost. Domain-specific search engines are becoming increasingly popular because they increase accuracy not possible with general, Web-wide search engines. Unfortunately, they are also burdensome and time-consuming to maintain. In this paper, we use machine learning techniques to greatly automate the creation and maintenance of domain-specific search. It describes new research in semi-supervised learning, text classification, and information extraction. We have built a demonstration system using these technics like Web Scrapping, Fuzzy C-Means and Hierarchy Clustering for a search engine which gives accurate results which is a more advantage when compared to other Search engines. Searching with a traditional, general purpose search engine would be extremely tedious or impossible to perform search operations. For this basis, domain-specific search engines are becoming popular. This article mainly concentrated on Project an effort to automate many aspects of creating and maintaining domain-specific search engines by using machine learning techniques. These techniques permit search engines to be created quickly with less effort and are suited for re-use across many domains.